SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Objective To determine the risk of recurrent cervical intraepithelial neoplasia (CIN) in women with complete or incomplete excision of cervical intraepithelial neoplasia treated by large loop excision of transformation zone (LLETZ).

Design A retrospective study

Setting One consultant-led colposcopy clinic at Leicester Royal Infirmary

Population Three hundred and ninety-four women referred consecutively to the colposcopy clinic between 1991 and 1992.

Main outcome measures The histological recurrence rate of CIN, length of cytological follow up following treatment related to degree of completeness of excision at initial treatment.

Results Three hundred and twenty-two women had complete cytological or histological follow up. The mean length of follow up was 73 months with a mean number of six smears. Women with incomplete excision of CIN had a significantly higher risk of recurrent CIN (relative risk 8.23) occurring in a significantly shorter time compared with women with complete excision.

Conclusions This study demonstrates that large loop excision of transformation zone is successful in over 95% of cases. Cytological surveillance is satisfactory for follow up of women who have complete excision of CIN. Women with incomplete excision of CIN at initial LLETZ remain at significant risk of developing further CIN and long term colposcopic and cytological follow up is necessary.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Large loop excision of the transformation zone (LLETZ) is the commonest technique for the outpatient treatment of cervical intraepithelial neoplasia (CIN)1. This technique, introduced approximately 10 years ago2, has gained in popularity, mainly because of its ease of use, acceptability to women and its low cost3. One of the main advantages of LLETZ over destructive treatments, such as laser ablation or cold cautery, is the availability of the excised specimen for histological diagnosis. This means that women may be seen and treated at one attendance at the clinic. The histological report should contain information on the grade of CIN and the com-pleteness of excision at the surgical margins.

Currently, the National Health Service Cervical Cancer Screening Program guidelines for follow up of women after treatment of CIN are a cervical smear six months following treatment and yearly thereafter for five years; if these smears are reported as normal the woman may return to the standard cervical smear screening programme4. Various local protocols are implemented for follow up of women who have undergone treatment of CIN. Some women will have incomplete excision of CIN after undergoing LLETZ. The treatment of these women is difficult as there are no guidelines. The aim of this study was to determine if women with incomplete excision of CIN by LLETZ were at increased risk of recurrent CIN compared with patients with complete excision and to determine if follow up of such women is safe by cytology alone.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Three hundred and ninety-four consecutive women, referred to one consultant-led colposcopy clinic between 1991 and 1992, underwent outpatient LLETZ for CIN. The original histology from the LLETZ procedure, subsequent cytological records and further histo-logical reports were traced. Women with invasive cervical cancer diagnosed at the time of the original LLETZ were excluded from the study.

All women underwent outpatient LLETZ using standard treatment protocols. If an identifiable area of CIN was noted at colposcopy, LLETZ was performed under local anaesthetic. Following LLETZ, ball diathermy was applied to the whole of the biopsy crater to establish haemostasis.

The histology reports stated the grade of the CIN and completeness of excision. If excision was incomplete, the site and extent of the incomplete excision was reported. After the treatment the standard practice was for cervical smears at six months and yearly for ten years thereafter. Women had their cervical smears performed by their general practitioner, and only returned for colposcopy if any smear was reported as abnormal or the cervix appeared abnormal.

Women were categorised into completely or incompletely excised CIN on the basis of the initial histology report. Incompletely excised cervical lesions were further subdivided into ectocervical, endocervical or mixed, depending on the site of incomplete excision. Any specimens with equivocal reports were examined by one pathologist (L.J.R.B.) and assigned to the appropriate group. Follow up cytology was traced using a computer list. Recurrence or persistence of CIN was determined by histological examination of further cervical biopsy or hysterectomy specimens.

Statistical analysis

The relationship of recurrence of CIN and complete or incomplete excision was assessed by the χ2 test. The times to recurrence in the complete and incomplete excision groups were compared by Student's t test.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Complete follow up data was available on 343 women (87%); there were no demographic or histological differences between the women lost to follow up (n= 51) and the women in the study. The majority of the women lost to follow up had moved away from the area. The age range of the women was 18–59 years. Two hundred and forty-six women (77%) had histological complete excision and 75 (23%) had incomplete excision. Eight women (2%) had incomplete excision at both endocervical and ectocervical margins; these women were excluded from the study as they underwent further LLETZ three months after the initial treatment. The adequacy of excision of 13 women (4%) was unclassified following review of the original histology, due to fragmentation at the margins of the histological specimens.

Of the women with incomplete excision, 37 (49%) had incomplete excision at the endocervical margins and 38 (51%) had incomplete excision at the ectocervical margins. The relationships between histological grade, age and excision margins are shown in Table 1.

Table 1.  Characteristics of the women undergoing outpatient large loop excision of transformation zone (LLETZ). Values are given asn or n (%), unless otherwise indicated. CIN = cervical intraepithelial neoplasia.
    Incomplete excision
Histology typeNo.Mean age (years)Complete excisionEctocervixEndocervixTOTAL
CIN I5632.946 (82)3710 (18)
CIN II9731.178 (80)71219 (20)
CIN III16834122 (73)271946 (27)
TOTAL321 246 (77)373875 (23)

Follow up data

Complete cytological and histological follow up data on 321 (82%) women were available. The mean length of follow up was 73 months (range 3–95 months) and the mean number of smears performed was six (range 0–11 smears). Fourteen women (4%) had histological recurrence of CIN. Within this group two women had developed invasive cervical carcinoma following original incomplete excision of CIN III. The invasive carcinomas were both Stage 1 A and were treated by simple hysterectomy.

There were only four women (2%) with complete excision who had recurrent disease, compared with 10 women (13%) with incomplete excision. Recurrence rates were statistically significantly different between completely and incompletely excised groups (2%vs 13%: P < 0.001) and in the time to recurrence (49 months vs 22 months: P < 0.01) (see Table 2).

Table 2.  Women with recurrent cervical intraepithelial neoplasia (CIN) following large loop excision of transformation zone (LLETZ): relationship to completeness of excision at initial treatment. There was a significant difference between the number of women with recurrent CIN and the mean times of recurrent CIN between complete excision and incomplete excision (1.6%vs 146%P < 0.001:49 months vs 22 months P < 0.001). Values are given as n or n (%), unless otherwise indicated. SCC = invasive squamous cell carcinoma.
  Incomplete excision
 Complete excisionEctocervixEndocervixTOTAL
Recurrence4/247 (2)5/375/3810/75 (13)
CIN I211
CIN II/III2437
SCC (stage)1 (1 AI)1 (1 AI)2
Time (months) (mean)49212422

Within the 10 women with recurrent or residual CIN and incomplete excision at time of the original LLETZ, five originally had incomplete excision at the ectocervi-cal margin and five originally had incomplete excision at the endocervical margin. Women with incomplete ectocervical margins needing further treatment had a further LLETZ on average 21 months later (range 4–71 months), while those with incomplete endocervical margins needing further treatment had further treatment on average 24 months later (range 8–32).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

This is the longest follow up study to date on women undergoing LLETZ. Our results reveal that the overall success rate of obliterating CIN by outpatient LLETZ was 308/321 (96%). This compares with reported success rates 89%–95%5,6 and is within the national guideline of 95% success rate. The results also indicate that incomplete excision of CIN by LLETZ results in a statistically significant increased overall recurrence rate compared with complete excision of CIN by LLETZ (13.33%vs 1.6%P < 0.001).

In our colposcopy clinic the standard practice in women with incomplete excision is cytological follow up by the general practitioner, unless there is involvement of both endocervical and ectocervical margins. Our results show that there is a significantly increased risk of recurrent CIN in women with incomplete excision, the relative risk being 8.23 (95% CI 3.19–21.6) These results suggest that women with incomplete excision of LLETZ should have colposcopic and cytological follow up.

Two women in this study developed invasive carcinoma (Stage 1 AI), having had previous incomplete excision of CIN III. In both women the microinvasive carcinoma was diagnosed incidentally and treated by simple hysterectomy. One woman with microinvasive carcinoma was diagnosed eight months after her initial treatment following an abnormal smear suggesting severe dyskaryosis, and the second woman had abnormal cytology suggesting severe dyskaryosis 71 months after her initial treatment. The risk of developing invasive carcinoma of the cervix following treatment of CIN is five times greater than the normal population7 and therefore justifies the need for continued cytological surveillance following treatment, even in women with complete excision of cervical intraepithelial neoplasia. All women with abnormal cytology following LLETZ should be referred immediately for colposcopic assessment of the cervix.

The rate of recurrence of CIN of 13% in women with incomplete excision contrasts with a recently reported recurrence rate of 31%8. One possible explanation for variations in recurrence rate maybe due to differences in technique in performing LLETZ. In our colposcopy clinic all women have ball diathermy to the LLETZ crater in order to prevent haemorrhage and to destroy a greater volume of cervix. Thus, if excision is incomplete some or all of the residual CIN will be destroyed by the ball diathermy. This theory is supported by Murdoch et al.6, who adopt a similar technique and reported a 95% cure rate despite a rate of incomplete excision of 44%. Incomplete excision margins at the time of LLETZ does not necessarily imply residual disease. Almost one third of women with residual or recurrent disease had a report of complete excision at initial LLETZ, possibly reflecting the increased risk that all women who are treated for CIN have for developing further CIN8.

In our study the majority of women with recurrent or residual disease following incomplete excision originally had CIN III excised (46/75 [61%])(Table 1). Women with CIN III and incomplete excisional margins are the most likely group to have recurrent or residual disease. This is not surprising as CIN III tends to be larger in area9 and extend into cervical crypts10, compared with lower grades of CIN. In our study eight out of 46 (17%) women with incompletely excised CIN III had recurrent or residual CIN detected at follow up, in agreement with other studies3,6,7.

There are no guidelines for the follow up of women with incomplete excision of CIN. Some authors recommend cytological follow up only11, others a second treatment12, and others combined cytology and colposcopic assessment of the cervix13.

Our study shows that the time to recurrence with complete excision was longer than with incomplete excision (49 months vs 22 months; P < 0.001). Incomplete excision may result in persistent disease, while complete excision may result in recurrent disease.

We conclude that cytological follow up alone is sufficient in women with complete excision of CIN by LLETZ. Colposcopic follow up should be performed in patients with incompletely excised CIN, as they have a greatly increased risk of persistent or recurrent disease (relative risk 8.23). This practice, however, will lead to an increased number of women undergoing cytological and colposcopic surveillance, an extra burden on the colposcopic services. In our practice this policy would have resulted in at least 65 extra colposcopic examinations. If this practice is generally adopted extra resources will have to be made available for colposcopy.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References